# Grade

> Use when adversarially grading a change for red proof, real behavior, cross-family independence, and live capability before shipping.

- Skill: `tcuzzo/grade` (Agent Skill)
- Install (CLI): `npx skillmds@latest add tcuzzo/grade`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tcuzzo/grade/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: Tcuzzo (https://skillmd.com/u/tcuzzo)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tcuzzo/grade

---


# Grade Command

Load and execute the complete [grading-verification play](../../plays/grading-verification.md)
on the current diff or the user's supplied scope. In a copied portable install, load
`~/.local/share/backs-aios/current/plays/grading-verification.md` instead. A green test
is a claim; keep attacking until the play's proof bar is met or the defect is loud.